Table 1. Confusion matrix used to evaluate the ability of the deep-learning algorithm to detect abnormal rock bream swimming behaviour

Abnormal class		Actual behaviour
		Abnormal	Normal
Predicted behaviour	Abnormal	25 (TP)	2 (FP)
	Normal	0 (FN)	78 (TN)

The detection of abnormal fish was described as true positive (TP), false positive (FP), true negative (TN), and false negative (FN) depending on the relationship between the actual measurements and predicted results.